ARAMIS (the American Rheumatism Association Medical Information System) is a rheumatic disease computer data bank system containing longitudinal clinical data for approximately 19,000 patients and 120,000 patient encounters, and representing more than 100,000 patient-years of observation. The system operates from an IBM 370/3081 computer at Stanford University and is accessed nationally through TYMNET or TELENET communications networks using the Time-Oriented Data Bank (TOD) data management system. The program is based upon the premises that chronic diseases have become the most prevalent health problems, that study of such diseases requires observation of occurrences over prolonged time periods, that the expense of longitudinal study requires use of economies of scale, that patient outcome in chronic disease results from a complex interplay between multiple factors, and that many important questions need to be studied with observational, in addition to experimental, techniques. This program has the goal of improving knowledge, management, and patient outcome in arthritis by providing long-term information relating disease severity, patient characteristics, social factors, and treatment to patient outcome. The program has two major aims; first to continue to develop a national data resource of high quality, longitudinal, accessible clinical data, and second, to employ these data in a systematic, multicenter investigative program of major clinical questions in the rheumatic diseases. Program priorities include the classification and definition of diseases, the systematic study of long-term (6 to 20 year) outcomes, the economic impact of illness and treatment, and study of regional and national differences. Thirty clinical investigators and epidemiologists at 12 institutions undertake over 50 projects annually. The present proposal includes classification studies of osteoarthritis, rheumatoid arthritis, vasculitis, and systemic lupus erythematosus, economic impact studies in each major disease, comparative studies of arthritis at different sites, population based studies of incidence and prevalence, and long-term studies of outcome in rheumatoid arthritis, juvenile arthritis, scleroderma, systemic lupus, osteoarthritis, and following joint surgery. With this project, 15 years of data development at numerous institutions are brought to bear upon major clinical questions, and very large and detailed longitudinal patient data sets are made nationally available.